Output

The flag -g can be used to indicate the output format of Rcross. In contrast to the input formats, there are seven options for output. Rcross will write output in a format suitable for Mapmaker/QTL if the -g option is used with the integer 2, while a cross.inp formatted file will be written with the value 1. Using a value of 3 produces a file suitable for input into Splus or R, while a 4 generates a SAS program file. The integers 5 and 6 tell Rcross to produce files suitable for import into PLABQTL [Utz and MelchingerUtz and Melchinger1996]: Use 5 if you want the matrix format and 6 for the vector format.

The default is what we term the qtlcart.cro format, and is indicated by using zero with the -g option. Here is an example of the output of Rcross.

#     1472574604   -filetype Rcross.out 
#
#	QTL Cartographer V. 1.12c, March, 1997
#
-n        300   is the sample size
-p         63   is one more than the number of markers
-cross   B1   is the type of cross
-traits     1   is the number of traits
-Names of the traits...
  1 Trait.1
-otraits    0   is the number of other traits
# 
-s
  1  1 
1 1 1 1 1 1 1 2 2 2 2 2 1 1 1 1 
2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 
2 2 2 2 2 2 2 2 2 2 2 2 2 1 
1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 
   7.035406650635
  2  1 
1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 
2 2 2 2 2 2 2 2 1 1 1 1 1 1 
1 1 1 1 2 2 1 1 1 1 1 1 1 1 1 1 
   3.555115422473
  3  1 
2 1 1 1 2 2 2 2 2 2 2 1 2 2 2 2 
1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 
2 2 2 2 2 2 2 2 2 2 2 1 1 1 
2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 
   4.165996548162
.
.
.
-e

The section prior to the `-s' token is self-explanatory. The area between the `-s' and the `-e' is the data. It starts with an identification number (1, 2, 3, etc), and is followed by a ``1''. At the moment this number ``1'' is ignored. With the convention that $A_1$ alleles originated from the $P_1$ line and $A_2$ from the $P_2$, marker genotypes will be encoded with the following integer values:

• 2 for $A_1 A_1$
• 1 for $A_1 A_2$
• 0 for $A_2 A_2$
• 12 for $A_1-$, that is individuals with at least one dominant $A_1$ allele
• 10 for $A_2-$, that is individuals with at least one dominant $A_2$ allele
• -1 for unknown genotypes. Rcross read something but could not translate it.
• -2 is also for unknown genotypes. In this case, no data had been read in.

The trait values follow the marker genotypes, and finally the ``other'' (categorical or qualitative) traits follow at the end. The sequence repeats for all individuals. Note that there is a permissible range for trait values. By default, all trait values must be real numbers with absolute value less than one million ($10^6$). Any trait value that is less than negative one million is treated as a missing phenotype by the programs.